Análisis multivariado y clustering de indicadores de siniestralidad vial en Ecuador: Un estudio 2020 – 2023

Jessica Andrea Vivanco Correa; Ciro Diego Radicelli-García

doi:10.37431/conectividad.v7i2.517

Authors

Jessica Andrea Vivanco Correa Universidad Nacional de Chimborazo https://orcid.org/0009-0000-5737-794X
Ciro Diego Radicelli-García Universidad Nacional de Chimborazo https://orcid.org/0000-0001-9188-0514

DOI:

https://doi.org/10.37431/conectividad.v7i2.517

Keywords:

Multivariate analysis, Road traffic accidents, Principal components, K-means, Ecuador

Abstract

In Ecuador, traffic accidents do not follow the same pattern across all provinces. In some cases, high volumes of accidents, population, and vehicles predominate; in others, the frequency is lower, but the relative fatality rate is higher. Based on this difference, the study analyzed road accident patterns during the 2020–2023 period. Data were taken from INEC and ANT records. The final database included 92 province-year observations, corresponding to four years of study and 23 provinces in Ecuador. The Galápagos Islands were excluded because some accident indicators had missing or very low values. Seven variables were analyzed: accidents, fatalities, injuries, registered vehicles, population, mortality rate, and accidents per vehicle. The data were first standardized using Z-scores, and then Principal Component Analysis was applied. The first two components explained 86.63% of the total variability: PC1 mainly represented the overall volume of road crashes and road exposure, PC2 was associated with relative severity, and PC3 with crash intensity relative to the vehicle fleet. Finally, the K-means algorithm was used to group the observations. The clustering identified three profiles: high absolute accident rate, intermediate performance, and higher relative severity. These results show that the combined use of PCA and K-means facilitates the identification of regional road risk profiles.

References

Acosta-González, N., Cahueñas, S., & Pérez, C. (2025). Risk factors for fatal road traffic accidents in Ecuador. Transportation Research Interdisciplinary Perspectives, 32, 101515. https://doi.org/10.1016/j.trip.2025.101515

Agencia Nacional de Tránsito. (2023). Estadísticas de siniestros de tránsito. https://www.ant.gob.ec/estadisticas/

Alam, M. S., & Tabassum, N. J. (2023). Spatial pattern identification and crash severity analysis of road traffic crash hot spots in Ohio. Heliyon, 9(5), e16303. https://doi.org/10.1016/j.heliyon.2023.e16303

Gedamu, W. T., Plank-Wiedenbeck, U., & Wodajo, B. T. (2024). A spatial autocorrelation analysis of road traffic crash by severity using Moran’s I spatial statistics: A comparative study of Addis Ababa and Berlin cities. Accident Analysis & Prevention, 200, 107535. https://doi.org/10.1016/j.aap.2024.107535

Instituto Nacional de Estadística y Censos. (2023). Estadísticas de siniestros de tránsito. https://www.ecuadorencifras.gob.ec/estadisticas-siniestros-de-transito/

Instituto Nacional de Estadística y Censos. (2024). Siniestros de tránsito: cuarto trimestre 2023. https://www.ecuadorencifras.gob.ec/documentos/web-inec/Estadisticas_Economicas/Estadistica%20de%20Transporte/2023/iv_trimestre/2023_RESULTADOS_SINIESTROS_IVT.pdf

Jolliffe, I. T., & Cadima, J. (2016). Principal component analysis: A review and recent developments. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 374(2065), 20150202. https://doi.org/10.1098/rsta.2015.0202

Kaufman, L., & Rousseeuw, P. J. (2005). Finding groups in data: An introduction to cluster analysis. Wiley. https://doi.org/10.1002/9780470316801

Li, J., & Chen, F. (2024). Gauging road safety advances using a hybrid EWM–PROMETHEE II–DBSCAN model with machine learning. Frontiers in Public Health, 12, 1413031. https://doi.org/10.3389/fpubh.2024.1413031

MacQueen, J. (1967). Some methods for classification and analysis of multivariate observations. En Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics: 5.1 (pp. 281–298). University of California Press. https://projecteuclid.org/ebooks/berkeleysymposium-on-mathematical-statistics-andprobability/Proceedings-of-the-Fifth-Berkeley-Symposium-onMathematical-Statistics-and/chapter/Some-methods-forclassification-and-analysis-of-multivariateobservations/bsmsp/1200512992

Maugeri, A., Barchitta, M., Favara, G., La Mastra, C., La Rosa, M. C., Magnano San Lio, R., & Agodi, A. (2023). The Application of Clustering on Principal Components for Nutritional Epidemiology: A Workflow to Derive Dietary Patterns. Nutrients, 15(1), 195. https://doi.org/10.3390/nu15010195

Muñoz Muñoz, E. G., Verduga Alcívar, D. A., Yandri Francinet, G. A., Lapo Palacios, M. A., & Zorrilla Briones, O. (2024). Búsqueda de patrones con machine learning en datos de siniestros de tránsito. Ciencia Latina Revista Científica Multidisciplinar, 8(2), 1617–1638. https://doi.org/10.37811/cl_rcm.v8i2.10592

Toalombo Vargas, V. M., & Montalvo Márquez, F. J. (2024). Severity of traffic accidents in Ecuador. Tesla Revista Científica, 4(1), e290. https://doi.org/10.55204/trc.v4i1.e290

World Health Organization. (2023). Global status report on road safety 2023. World Health Organization. https://www.who.int/publications/i/item/9789240086517

Yang, W., & Han, S. (2025). Direct and indirect effects of road attributes on traffic safety. Journal of Safety Research, 93, 156–169. https://doi.org/10.1016/j.jsr.2025.02.017

Zeng, Y., Qiang, Y., Zhang, N., Yang, X., Zhao, Z., & Wang, X. (2024). An Influencing Factors Analysis of Road Traffic Accidents Based on the Analytic Hierarchy Process and the Minimum Discrimination Information Principle. Sustainability, 16(16), 6767. https://doi.org/10.3390/su16166767

Zhang, X., Qi, S., Zheng, A., Luo, Y., & Hao, S. (2023). Data-driven analysis of fatal urban traffic accident characteristics and safety enhancement research. Sustainability, 15(4), 3259. https://doi.org/10.3390/su15043259